The gravity center position of truck crane boom is necessary for boom structural design, crane load capacity setting and counterweight design. The position of the boom's gravity center is affected by the deflection caused by gravity. Therefore, firstly, the segment by segment hardening and superposition method is adopted to derive the analytical formula of the boom's deflection, and then the center of gravity coordinates is calculated according to the boom's structure and deflection solution. The center of gravity coordinate calculation program of crane boom is compiled by using Microsoft Visual Studio, and the it is carried out by taking the 5-section boom of XCT25L5 crane as an example. The comparison with the finite element analysis results shows that the algorithm in this paper is satisfied, and the average error is within 10%. In this study, a simple and practical calculation formula of boom center of gravity coordinates is derived through theoretical analysis, and the rapid calculation of boom center of gravity coordinates is realized through software, which can be used in various R & D designs and reduce the development cycle and cost.
In the intelligent fault diagnosis of rolling bearings, the high recognition accuracy is hardly achieved when small training samples and strong noise happen. In this article, a novel fault diagnosis method is proposed, that is radial basis function neural network with power spectrum of Welch method. This fault diagnosis model adopts the way of end-to-end operating mode. It takes the original vibration signal (time-domain signal) as input, and Welch method transforms the data from time-domain signals to power spectrums and suppresses high strength noise. Then the results of Welch method are classified by radial basis function neural network. To test the performance of radial basis function neural network with power spectrum of Welch method, the method is compared with some advanced fault diagnosis methods, and the limit performance test for radial basis function neural network with power spectrum of Welch method is carried out to obtain its ultimate diagnosis ability. The results show that the proposed method can realize the high diagnostic precision without the complex feature extraction from the signal. At the same time, in the case of a small amount of training data, this method also can achieve the diagnosis in high precision. Moreover, the anti-noise performance of radial basis function neural network with power spectrum of Welch method is better than the performance of some fault diagnosis methods proposed in recent years.
In robotics, robot forward and inverse kinematics and dynamics are the foundation of robot research. Aiming at the special working environment and wafer falling off of mag8 vacuum manipulator produced by Shenyang SIASUN Robot Automation Co., Ltd., this paper completes the research on motion planning and compliance control of the manipulator. Firstly, its mechanical structure and motion mode are briefly introduced, which can realize the lifting, rotation and telescopic actions of both arms. Secondly, the robot kinematics model is established by MDH method, and the robot dynamics model is established by Lagrange method. Finally, experiments are carried out through MATLAB robot toolbox to verify the correctness of forward and inverse kinematics and dynamic model. The experimental results show that the dual arm model established in this paper effectively simulates the actual working state, which lays a foundation for further physical platform operation.
In many industrial fields, the triangular mesh had replaced the original CAD models for their simplicity and efficiency. The mesh surfaces have been widely used in the direct machining for the industrial needs. Compared with the general machining methods, the multi-axis machining can achieve the better curved surface processing effect. However, the tool axis vector cannot be accurately obtained from the mesh surface, and the multi-axis NC machining cannot be implemented directly and effectively. Base on the method that the tool axis vector perpendicular to the part, this paper proposes use the normal vector of the facet to approximatively substitute the tool axis vector and achieve the multi-axis NC machining of the mesh surface. To obtain the normal vector of the facet, the operations include facet data acquisition, regenerated triangles area calculation, selected facet judgement and extract the normal vector are introduced. And the proposed method is verified by computer programming and running in IDEL. This method will be benefit of the multi-axis NC machining of the mesh surface and solve the problem of tool axis vector generation.
To further improve the micro texture position tracking precision machine tools,the establishment of ball screw mechani-cal transmission model and dynamic model design of servo control system based on model error compensation scheme for the de-sign of the control system,puts forward a search algorithm based on improved crowd feedforward PID control strategy,estab-lished the ISOA-PID controller of the control structure,completed the design of PID parameter.Matlab is used to build a simula-tion experiment platform.In order to verify the actual effect of the control strategy,Matlab was used to build a simulation experi-ment platform and compared with the traditional intelligent optimization algorithm.The simulation results show that the feedfor-ward compensation PID optimization effect based on ISOA optimization is better,and the speed tracking performance is greatly improved,which verifies its correctness and effectiveness.
To solve the problems of incomplete information extraction and insufficient depth in the process of extracting the assem-bly feature information of complex 3D CAD assembly models,such as aircrafts、large conveyers which have many assembly parts and complex coordination,a method of extracting assembly feature information of complex products was proposed,which took CATIA software as the platform.Component Application Architecture(CAA)is used to develop the function of quick extraction of feature information.Combined with the idea of multi-tree pre-sequence traversal and the depth-first search algorithm,it can completely extract the fit and constraint information of multi-layer nested complex assembly,make full use of the feature informa-tion of complex3D CADassembly model,and more strongly support the assembly process planning.At the same time,the feasibil-ity of this method is verified by taking a civil aircraft wing conveyor as an example.
The current design method for wind turbine airfoils and vortex generators (VGs) is independent, which cannot be used to simultaneously design wind turbine airfoils and VGs with high aerodynamic performance. Therefore, an aerodynamic-shaped integrated design method for wind turbine airfoils and VGs is proposed. The aerodynamic-shaped VGs instead of plate VGs are introduced to be installed on wind turbine blade section. The blade airfoil and aerodynamic-shaped VGs are both expressed by B-spline function. Then, the optimal mathematic model of wind turbine airfoil with aerodynamic shaped VGs is established, for which the multi-objective functions are maximum lift to drag ratio plus maximum lift coefficient. DU97-W-300 and CLARKY-117 airfoils are selected as initial objects to be optimized simultaneously by combining CFD simulation and particle swarm optimization (PSO) algorithm. The optimal results indicate that the maximum lift coefficient and lift to drag ratio of the optimized blade with novel VGs show the increase of 9.3% and 7.5%, respectively, compared to initial design. In addition, through the comparative analysis of streamlines and vorticity contours, the new blade section with novel VGs could produce larger induced vortex and could effectively restrain the fluid separation on the blade surface. This study provides a good reference to the design and application of wind turbine airfoils and VGs.
针对Robocup等比赛中机器人的定位中的视觉特征选取问题,线特征以其抗噪性好,蕴含丰富环境信息的优势获得了人们的青睐.该文针对仿人足球机器人球场比赛定位中的边线特征提取,将整体方案分解为颜色空间映射、可能片段查找、增量式算法拟合三个部分.通过预先学习标定生成的颜色查找表,将原始图像映射成多值图像;之后针对不同颜色对应的不同状态,使用有限状态机方法识别可能属于边线特征的有效片段;使用增量式算法提取出图像中的直线特征,最后增加参数调节来进一步优化特征提取结果,提高直线特征的精度.这一方法不仅可以提取直线特征,同时对于不规则的曲线特征检测,也能够使用近似的线段集合进行拟合.相比于传统方法如Hough变换等,该算法具有算法简单,提取速度快,特征精度高的优点.实验表明,算法的全局定位精度在10cm以内,特征提取时间不大于5ms,符合足球机器人实际比赛需求.
Since its small turning radius and simple structure,the articulated vehicles are widely used in underground roadway transportation. Under in-situ steering,the phenomenon of“bending down”appears in this kind of vehicle here,and there are many influencing factors and large errors in the analysis model. According to the structural layout characteristics of the dual hydraulic cylinder steering system,the geometric analysis of the in-situ steering condition was carried out to obtain the steering trajectory of the condition. Based on the trajectory analysis,the analysis model was established by using Simulink. For the whole vehicle and the hydraulic system,the model was established by using Adams and AMESim,and the analysis model was established by combining the models. The force fitting equation of the hydraulic cylinder in the in-situ condition was applied to the model. The force on the hinge point and the track of the whole vehicle was analyzed. The change of the mass center position of the car body forward and backward was selected,and the influence of the force on the hinge point,the force on the wheel and the track was analyzed. The results show that:when turning in-situ steering,the movement direction of the inner and outer wheels of the front and rear car body is opposite,and the force direction is opposite,so the whole car can turn smoothly. The vertical force of the rear car body wheel fluctuates frequently,which has a great impact on the steering stability of articulated car. When the mass center of the rear car body moves forward or backward,the longitudinal and transverse forces on the articulated body increase,and when the rear car body moves backward,the force situation is more serious. The results show that the in-situ steering trajectory obtained by the model is similar to the“bending phenomenon”,which is consistent with the actual operation,solves the problem of large model error,and provides a reference for such design and research.
在汽车稳态转向性能开发过程中,工程师会关注一些稳态指标,如不足转向度,侧倾梯度,方向盘力矩等,但较少注意得到这些指标的试验方法和数据处理方法.阐述了对于测试汽车稳态转向特性不同试验方法的理解,对比了定半径、定车速以及定转角方法的效率与难度,以及侧向加速度不同获取方法的差异,不同设备测试侧倾角的差异,不同试验方法得到的不足转向度、侧倾梯度以及方向盘力矩的差异.建议在稳态转向性能开发时,利用相同的试验方法和数据处理方法对比不同汽车的性能.
针对大型装备动态变形的三维测量难题,以不同风速下风力发电机叶片的动态特性为基础,提出了一种基于双目立体视觉与数字图像相关法相结合的动态测量方法,对风力发电机叶片进行了动态测量.采用工业摄影相机,对旋转叶片的三维全尺寸轮廓快速拍摄,通过测量原理和光学扫描方法,快速解算出被测叶片的三维坐标和变形应变.以试验样机叶片为例,对不同风速的旋转叶片进行全尺寸动态应变检测试验,并与有限元分析结果、光纤光栅传感器检测结果进行对比验证.结果表明:三者结果基本一致,验证了该检测方法的有效性与合理性.
针对滚动轴承保持架由于故障频率太小容易被噪声干扰,振动分析等传统故障检测方法检测困难,特征较难提取的问题,提出一种基于迁移学习的轴承保持架的故障诊断算法.利用数据量较多但缺少保持架故障相关数据的凯斯西储大学的轴承振动加速度数据集进行模型训练,提取出重要的模型参数信息,然后在此基础上,利用此模型参数在少量的齿轮箱轴承保持架振动加速度数据上进行迁移学习,实现对齿轮箱轴承保持架的故障识别,实验表明该迁移学习方法在识别滚动轴承保持架上是有效的.
电传动可有效提升短距离运输自卸车的载重量,而被广泛应用于此类车辆.针对电传动自卸车交流驱动系统的特点,提出交流传动控制系统结构设计方案.以数字处理器DSP与CPLD为核心的电路方案,并提出以功率为外环的矢量控制策略来控制异步牵引电机,对控制程序进行设计;根据现有电机选择系统中所需要的各种器件,设计各种保护与检测电路;采取电阻消耗式交-直-交驱动方案,搭建牵引异步电机试验平台,分析母线电压升高对电机影响,对车辆满载停车再次启动工况等工况进行测试.试验结果表明所设计的系统能够保证输出电流不过流;无电流冲击及振荡现象产生,能够适应负载的各种变化;满足电传动矿用自卸车在实际运行中对路况复杂多变的要求.为此类车辆控制系统设计提供参考.
探究液体磁性磨具对伺服阀阀芯光整效果的影响,找到相关因素对光整加工效果的影响规律,为阀芯的液体磁性磨具光整加工在实际生产上的应用提供指导.对阀芯的液体磁性磨具光整加工进行分析,提取了影响阀芯光整效果的主要因素并进行单因素加工实验.一定范围内,阀芯表面粗糙度下降率%ΔRa随工件转速、磨料目数和磨料质量分数的增大而增大,而Z向振动频率对表面粗糙度的变化影响小;阀芯均压槽棱边毛刺被完全去除的时间随着工件转速、振动频率和磨料质量分数的增大而缩短,随磨料目数的增大而增长.工件转速为1400r/min、Z向振动频率为2.5Hz、磨料目数为80#、磨料质量分数为25%时,光整10min后,表面粗糙度Ra由0.27μm下降到0.155μm,%ΔRa为38.89%,阀芯棱边毛刺被完全去除.阀芯的液体磁性磨具光整加工中,工件回转运动为降低Ra的主运动,工件回转运动和直线振动在毛刺去除上均起主要作用.光整后阀芯Ra明显下降,棱边平整光滑,均压槽内清洁度和均压槽底光亮度均有所改善,光整效果良好.
针对传统局部路径规划存在非全局最优、易陷入困境、导航效率低等问题,这里提出了一种将改进RRT(Rapidly-Exploring Random Trees)算法和动态窗口法融合的算法.首先优化基本RRT算法的采样策略,使用三次贝塞尔曲线平滑所生成的全局路径.然后改进DWA(Dynamic Window Approach)算法的轨迹评价函数,构造路径最优的目标函数,以保证路径规划最优,从而提高移动机器人的避障性能.最后使用ROS(Robot Operating System)平台进行仿真验证,实验结果表明,与A*-DWA和Dijkstra-DWA相比,所提出算法在复杂环境下路径长度更短、路径质量更优,行进时间明显减少,移动机器人的平均速度较A*-DWA算法提高了约16.8%,证明了算法的有效性和实用性.
在一些高新技术装备中有一类轴孔零件:体积大、质量重,轴孔无倒角,且配合间隙为0.05mm左右.这类零件的装配多以人工手动操作为主,但手动装配较难控制装配接触力及配合精度,在装配过程中又存在较高的磕碰风险.这里对轴孔高精度装配方法及精度进行了分析,提出了基于视觉+力位混合控制的柔顺装配方法.并设计了轴孔装配实验系统,对直接装配、示教装配、柔顺装配方法开展了实验研究,通过实验发现:由于视觉测量的精度较低,直接装配方法无法实现精密装配;由于机器人重复精度的影响,基于示教的重复装配方法在多次装配后无法正常入装;基于视觉+力位反馈,柔顺控制算法得到调节量,机器人在5次左右调节后可完成轴孔的精密装配.
为了准确对巡检机器人穿越预定杆塔所需要的能量进行SOC估计,通过分析线路工况和实验数据,建立巡检机器人能耗模型.考虑到巡检机器人从起始杆塔到预定杆塔之间的累计误差会逐渐增大,导致抵达预定杆塔时无法对巡检机器人锂电池SOC准确估计.因此,结合巡检机器人能耗状态方程和锂电池量测模型,并采用扩展卡尔曼滤波的方法对巡检机器人锂电池SOC能耗的理论值进行迭代,同时,也引入次优渐消因子降低不确定参数的误差,提高SOC估计精度,且SOC估计误差均在1.7%以下.
为了提高激光SLAM导航定位的实时性和稳定性,采用自适应阈值法、迭代适应点以及最小二乘法结合直线度的方法提取局部地图中的特征线段和全局地图中的线段,通过特征线段匹配进行初始定位.利用惯性导航位姿推算结合地图匹配的方法进行动态定位,获得机器人的实时位姿.同时,采用动态重定位的方法进行重定位,提高了移动机器人对工作环境的适应能力和可靠性.实验表明,基于地图匹配的导航定位算法的定位精度在±40mm以内,定位的时间不大于0.03s,可以较好地满足机器人的实际导航需求.
针对倒立摆(IP)系统中存在的参数不确定和外部扰动问题,设计了一种新的基于状态误差端口受控哈密顿(EPCH)的控制方法.基于开环端口受控哈密顿(PCH)模型,利用互联配置、阻尼注入和能量成形(ES)的方法,构造了IP系统闭环EPCH模型.针对IP系统中存在的参数不确定,采用自适应方法设计了自适应参数控制律.通过状态扩展的坐标变换方法,提出了一种新的PI控制律,抑制系统中出现的外部扰动.所提出的控制律都保持了闭环系统的PCH结构,保证了系统渐近稳定.仿真和实验结果表明,所提出的控制策略与PID控制方法相比,系统的稳态误差更小、抗扰动能力更强、鲁棒性更好.
民用飞机机头设计是飞机设计的亮点和难点,流线型机头设计已然成为一种趋势.由于其约束条件复杂、型号进度紧张,设计难度较大.这里研究了一种参数化机头设计方法,区别于传统参数化,本方法不是简单直观地定义产品外形参数,也不是仅依靠单一约束来评判设计好坏,而是采用系统工程理念,全面捕获各类需求并综合权衡,聚焦于在复杂约束中提炼并自定义影响飞机整体性能的参数,形成了一套完整的优化设计方法.使用本方法设计出的机头外形方案通过了CFD计算和风洞试验选型验证,确认了该方法的可行性,极大地缩短了机头优化选型时间.
